Dynamic rupture of crosscutting faults: A possible rupture process for the 2007 M_w 6.6 Niigata‐ken Chuetsu‐Oki earthquake

Abstract

[1] The 2007 M w 6.6 Niigata-ken Chuetsu-Oki earthquake implies a complex fault mechanism. Several surveys suggest the possibility of two main segments crosscutting each other at their center (i.e., a northern segment dipping to the northwest and a southern segment dipping to the southeast). We modeled the dynamic rupture propagation numerically along the inferred segmented fault system using a boundary integral equation method (BIEM). The possibility of rupture transfer is numerically demonstrated, and two rupture modes are described. Simultaneous rupture transfer along the overlapping part is possible only under a high-stress load; however, this rupture mode yields an excessively high slip value. Otherwise, where regional stress is relatively low but pore pressure is high enough to govern the rupture criterion (described as the low frictional coefficient case in this study), the rupture transfer to the other segment does not occur until rupture terminates on the first segment regardless of the crosscutting distance between the two segments. In this case, at the central crosscutting part, a possible rupture scenario is that ruptures occur sequentially in different directions: first a southbound rupture on the northwest-dipping segment, then a northbound rupture triggered on the southeast-dipping segment a few seconds later. The simulated rupture scenario on crosscutting faults can be strongly influenced by a preexisting fault structure determined by Miocene rifting during an opening stage of the Sea of Japan and the subsequent shortening of the crust. Our results point to the importance of investigating earthquake rupture scenarios in complex fault systems based on existing geophysical and geological information and dynamic rupture mechanics.Citation: Aochi, H., and A. Kato (2010), Dynamic rupture of crosscutting faults: A possible rupture process for the 2007 M w 6.6 Niigata-ken Chuetsu-Oki earthquake,

Finite Difference Simulations of Seismic Wave Propagation for the 2007 Mw 6.6 Niigata-ken Chuetsu-Oki Earthquake: Validity of Models and Reliable Input Ground Motion in the Near-Field

“…Many kinematic inversions adopt the southeast dipping fault geometry, but this orientation is not the only one sustainable. Aochi and Kato (2010) demonstrate the possibility of a dynamic rupture transfer from a northwest dipping sub-fault to another southeast dipping sub-fault. The two models we used in this study differ in fault orientation, in the position of asperities and in rupture timing, and actually do not fit the same stations.…”

“…We approximated this effect numerically by means of 5 x 5 sub-sub-point sources distributed on each sub-fault in our finite difference simulation, as we already demonstrated in other earthquakes (e.g. Aochi et al, 2010). For this earthquake, we observe that the moving source effect is also visible at NIG004.…”

“…However on the basis of aftershock observations obtained from arrays deployed for one month after the mainshock, the interpretation that predominantly emerges is that the major part of faulting is southeast dipping (Kato et al, 2009). Multisegment models are inferred from the InSAR observations (Aoki et al, 2008;Nishimura et al, 2008), and the possibility of rupture transfer between multisegment models is also dynamically simulated (Aochi & Kato, 2010). Thus understanding this earthquake is a seismologically challenging problem, and it is worth ascertaining how well we can reproduce the ground motion using the known seismological information for this earthquake.…”

A Probable Earthquake Scenario near Istanbul Determined from Dynamic Simulations

“…On the other hand, numerical simulations incorporating the mechanics of earthquake rupture (hereafter termed dynamic simulation) revealed that the earthquake rupture, under typical tectonic loading, is controlled by intrinsic features of the fault macroscopic structure . In particular, the fault geometry plays a significant role in the earthquake process as inferred from numerical simulations (e.g., Harris and Day, 1999;Oglesby et al, 2003;Aochi and Kato, 2010;Kase, 2010;Fukuyama and Hao, 2013;Douilly et al, 2015), as well as from geological observations (e.g., King and Nábĕlek, 1985;Wesnousky, 2008). The aim of this study is to make use of high-performance computing of dynamic simulations to infer the likelihood of several earthquake scenarios in the Sea of Marmara.…”

Strong localized asperity of the 2011 Tohoku-Oki earthquake, Japan

“…We define the yield stress σ s = µ s σ e n , where µ s is the static friction coefficient and σ e n is the effective normal stress, and the dynamic stress σ d = µ d σ e n , where µ d is the dynamic friction coefficient. If we assume that σ f ∼ = σ d , σ s −σ i = σ i −σ d , and µ d = 0.5µ s (Aochi and Kato, 2010), we obtain the relation σ = 0.25µ s σ e n . If one assumes µ s = 0.7 as inferred from laboratory experiments (e.g., Byerlee, 1978), σ = 40 MPa yields σ e n = 230 MPa, which is consistent with the effective normal stress inferred by Kato and Yoshida (2011).…”